Biological Conservation 233 (2019) 1–11
Contents lists available at ScienceDirect
Biological Conservation
journal homepage: www.elsevier.com/locate/biocon
Review A tendency to simplify complex systems T ⁎ Robert A. Montgomerya, , Remington J. Molla, Elise Say-Sallazb, Marion Valeixb,c,d, Laura R. Prughe a Research on the Ecology of Carnivores and their Prey (RECaP) Laboratory, Department of Fisheries and Wildlife, Michigan State University, East Lansing, MI 48824, USA b Laboratoire de Biométrie et Biologie Evolutive, UMR 5558, Centre National de la Recherche Scientifique (CNRS), Université Claude Bernard Lyon 1, Bât Gregor Mendel, 43 Boulevard du 11 Novembre 1918, 69622 Villeurbanne Cedex, France c LTSER France, Zone Atelier “Hwange”, CNRS 15 HERD (Hwange Environmental Research Development) Program, Hwange National Park, Bag 62, Dete, Zimbabwe d Wildlife Conservation Research Unit, Department of Zoology, University of Oxford, Recanati19 Kaplan Centre, Tubney House, Abingdon Road, Oxford OX13 5QL, United Kingdom e School of Environmental and Forest Sciences, University of Washington, Seattle, WA 98195, USA
ARTICLE INFO ABSTRACT
Keywords: Predation is a fundamental force exerting strong selective pressure on prey populations. Predators not only kill Carnivore prey, triggering lethal effects, but also hunt prey which can induce risk effects. Foundational research has Lethal effects documented the importance of risk effects in predator-prey systems of arthropods, fish, birds, and rodents, ff Predator-prey, risk e ects among others. Risk effects research in carnivore-ungulate systems has expanded in the last 20 years. Presently, Trophic ecology the degree to which this research mirrors the complexity of carnivore-ungulate trophic systems has been Ungulate questioned. We synthesized this literature to quantify the tendency of risk effects research in carnivore-ungulate systems to be multispecies in design. Among the 170 studies that we reviewed, we found that on average just 1.26 (range = 1 to 5) carnivore species and 1.60 (range = 1 to 11) ungulate species were considered per study. Furthermore, 63% (n = 107 of 170) of the studies featured single predator - single prey research designs. These results contrast with the fact that all but one of the 82 carnivore-ungulate systems used this literature had multiple species of carnivores and/or ungulates. Thus, we detected a tendency to simplify complex systems. We relate these observations to the role of simplicity as: i) an underlying value of science (i.e., Occam's razor), ii)a cornerstone of predator-prey theory (e.g., Lotka-Volterra equations), and iii) part of the origins of risk effects research (i.e., experimental systems). Finally, we ground our discussion in the implications of this research for the conservation of carnivores and ungulates in the dynamic 21st century.
1. Introduction broadly, the general structure of trophic systems (Lindeman, 1942; Kerfoot and Sih, 1987; Cohen et al., 2003; Jonsson et al., 2005; Brose Trophic systems are a complex arrangement of biotic and abiotic et al., 2006; Barnes et al., 2010; McCauley et al., 2018). Furthermore, components that dynamically interact among and across perceived le- whether these systems are dictated by bottom-up processes, top-down vels of organization (Hairston et al., 1960; Polis and Strong, 1996). The processes, or, more likely, some combination of the two, governs the flow of nutrients and energy through these systems, coarsely consisting ways in which the effects of the interspecific interactions can cascade of primary producers as well as first-, second-, third-, and fourth-order through the system (Paine, 1980; Pace et al., 1999; Schmitz et al., 2000; consumers, has been called the Eltonian pyramid in acknowledgement of Finke and Denno, 2004). In recognition of the tremendous importance Charles Elton's pioneering research in the early 20th century (see Elton, of these interactions on the natural world, research assessing the 1927). It is within this pyramid of numbers, ecological pyramid,orfood principles of trophic ecology has been extensive (for reviews see Polis web that the interactions of floral and faunal species play out in a et al., 1997; Estes et al., 2011; Layman et al., 2015). producer-consumer paradigm (Pimm, 1982; Cohen and Newman, 1985; Importantly, the mechanisms that underlie the interactions of con- Winemiller and Polis, 1996). These interactions have subsequent im- sumers and producers (i.e., predators and prey) within trophic systems plications for animal allometry, species population sizes, and, more are not exclusively predicated upon direct predation. Lethal effects
⁎ Corresponding author. E-mail addresses: [email protected] (R.A. Montgomery), [email protected] (R.J. Moll), [email protected] (E. Say-Sallaz), [email protected] (M. Valeix), [email protected] (L.R. Prugh). https://doi.org/10.1016/j.biocon.2019.02.001 Received 15 December 2018; Received in revised form 15 January 2019; Accepted 1 February 2019 0006-3207/ © 2019 Elsevier Ltd. All rights reserved. R.A. Montgomery, et al. Biological Conservation 233 (2019) 1–11
(also known as consumptive effects) occur when predators kill prey calls among the scientific community have questioned the extent to (Paine, 1966; Taylor, 1984; Sih et al., 1985) and are integral to the which research on risk effects in carnivore-ungulate systems adequately functioning of trophic systems. However, predators also influence represents the complexity of these trophic systems (Cresswell and trophic systems along nonlethal pathways by inducing phenotypic Quinn, 2013; LaManna and Martin, 2016; Moll et al., 2017; Say-Sallaz changes in prey. While physiological, morphological, and behavioral et al., this issue). The trophic dynamics of most carnivore-ungulate modifications may reduce the predation risk experienced by prey, they systems are highly complex, where predation risk experienced by a often come at a cost (Lima and Dill, 1990; Abrams, 1995; Brown et al., number of ungulate prey species dynamically derives from one or more 1999; Tollrian and Harvell, 1999; Peacor and Werner, 1997, 2001; of several sympatric carnivore species. Consequently, what is needed is Schmitz et al., 2004; Creel and Christianson, 2008; Heithaus et al., studies from natural systems in which the complexity of multi-predator 2008). For example, prey might respond to prevailing predation risk by effects is assessed (Cresswell and Quinn, 2013; LaManna and Martin, increasing vigilance. This behavioral adjustment however, would cor- 2016; Creel et al., 2017; Northfield et al., 2017). In the absence of respondingly reduce foraging effort. These costs of induced anti- multiple predator – multiple prey studies, it will be difficult to de- predator behavior are typically referred to as nonconsumptive (Brown termine whether the underlying predator-prey frameworks are applic- and Kotler, 2004; Peckarsky et al., 2008; Peacor et al., 2013)orrisk able (Peckarsky and McIntosh, 1998; Thaker et al., 2011; Dröge et al., effects (Creel and Christianson, 2008; Heithaus et al., 2008). Though 2017; Schmitz et al., 2017). lethal effects have historically received the bulk of the research atten- Given this context, it is crucial to identify the degree to which risk tion in trophic ecology, the nature and strength of risk effects can be effects research in carnivore-ungulate systems has included multiple equivalent to, or even greater than, lethal effects (Schmitz et al., 1997; species of predators and prey. We hypothesize that risk effects research Creel et al., 2008; Pangle et al., 2007; Cresswell, 2008; Ford and in carnivore-ungulate systems has tended not to assess multispecies Goheen, 2015; Creel, 2018). Furthermore, the risk effects associated dynamics. In pursuit of this research hypothesis, we conducted an ex- with the decisions of individual prey can scale to have population-level tensive survey of the peer-reviewed literature. We ground the discus- consequences (Mangel and Clark, 1988; Lima, 1998, 2002; Sih et al., sion of the implications of this research in the origins of predator-prey 1998). theory and we highlight the ways in which multi-species experimental Foundational research on trophic ecology and lethal and risk effects designs may yield novel and original insights in carnivore-ungulate has typically been associated with relatively small experimental sys- ecology. Our analysis has important implications for conservation be- tems featuring relatively small (e.g., < 1 kg) species (see Schmitz et al., cause overly simplistic research approaches will provide outputs that 2017). Take for example the research on optimal foraging theory con- are too general or too inaccurate to have a meaningful impact. This is ducted by Werner and Mittelbach (1981) “in a small Michigan lake and particularly important in carnivore-ungulate systems given the rates at an artificial pond” (p. 820) of predatory fish between 101 and 150 mm which large carnivores and their ungulate prey are declining the world in length. Equally formative work on risk effects has been conducted over. Thus, we comment on the ways in which simplicity in assessing among systems featuring predators pursuing prey comprised of insects risk effects among carnivores and their ungulate prey might hinder the (Schmitz et al., 1997; Schmitz, 1998), snails (Turner, 1996; Bernot and development of progressive policies meant to conserve these animals in Turner, 2001; Turner and Montgomery, 2003), birds (Cresswell, 1993, a dynamic world. 1994), rodents (Brown et al., 1988; Kotler, 1984; Brown and Kotler, 2004), fish (Werner et al., 1983; Turner and Mittelbach, 1990), and 2. Methods turtles (Heithaus et al., 2002, 2007). Much has been learned via this work, but the extent to which the principles are generalizable to larger 2.1. Literature review species inhabiting vaster terrestrial systems has been called into ques- tion (see Pearson and Dawson, 2003; Ricklefs, 2008; D'amen et al., We conducted an extensive review of literature (completed in July 2017; MacLeod et al., 2018). In addition to the focus on smaller or- 2018) evaluating risk effects in carnivore-ungulate systems. We con- ganisms, the origins of predator-prey theory tended to focus on simple ducted this review in the Web of Science search engine using the fol- systems or trophic structures featuring single predator-prey species lowing terms: (carnivore AND ungulate) AND (risk effects OR non- couplings (e.g., Lotka, 1925; Volterra, 1926; Nicholson and Bailey, consumptive OR predation risk OR nonlethal OR non-lethal OR trait- 1935). mediated OR behaviorally-mediated OR landscape of fear). Our next The context provided above helps explain why, until recently, re- step was to assess all literature deriving from this review. We retained search on risk effects has not typically been situated in systems com- those studies featuring objectives statements that were consistent with prised of carnivores pursuing mobile and elusive ungulate prey. For our analysis and eliminated from consideration any unrelated studies. example, among a meta-analysis of risk effects in predator-prey inter- Unrelated studies included those that did not center their assessment on actions published in 2005, only one of the 453 studies reviewed as- risk effects or those that assessed risk effects in systems not including sessed carnivores and ungulates (see Preisser et al., 2005). Risk effects carnivores and ungulates. It is also important to note that our interest research in carnivore-ungulate systems has likely lagged behind that here was in predation rather than depredation. Depredation refers to conducted in smaller systems given important logistical, technological, instances in which predators hunt domestic animals including cats (Felis and ethical challenges of studying large and charismatic species (see silvestris catus), dogs (Canis familiaris), and a variety of livestock and Estes, 1995). Over the last 20 years however, research on risk effects in poultry (Ogada et al., 2003; Woodroffe et al., 2005). Given that de- carnivore-ungulate systems has greatly increased (see Moll et al., 2017; predation was not relevant to our assessment, the studies that we Say-Sallaz et al., this issue). The rapid expansion in this research may be evaluated in this review focused exclusively on predation risk effects on demonstrative of unsustainable growth as considerable methodological wild ungulates. Among the resultant set of literature, we recorded: i) variation among the carnivore-ungulate risk effects literature has been the study objective, ii) the ecosystem in which the study was positioned, observed (Moll et al., 2017). This has called into question the com- and both the iii) predatory carnivore species and iv) ungulate prey parability of this research across study sites, leading to subsequent re- species evaluated in each study. commendations to standardize the methods used in this research (see Moll et al., 2017; Prugh et al., 2019). 2.2. Trophic interactions studied Such standardization is important given that it is these systems with attacking carnivores and adept ungulate prey that present ideal can- We evaluated each of the studies to identify the trophic interaction didates for research examining the nature and strength of risk effects (s) assessed. We grounded these interactions within the level of the (Brown et al., 1999; Creel et al., 2008; Schmitz et al., 2017). Recent trophic system involving predation between species in the order
2 R.A. Montgomery, et al. Biological Conservation 233 (2019) 1–11
Carnivora and infra-order Ungulata. As the nature and strength of risk Roemer et al., 2009). Thus, the apex and mesopredator designations effects has been found to vary according to predator hunting mode (see that we made are trophic system specific (see Prugh et al., 2009; Ritchie Preisser et al., 2007; Miller et al., 2014; Schmitz, 2008; Schmitz et al., and Johnson, 2009; Fleming et al., 2017; Haidir et al., 2018). There 2017) we subdivided Carnivora by hunting mode. Classically, there are were a couple of species featured in this literature review (wild dogs - three hunting modes (active, sit-and-wait, and sit-and-pursue) by which Lycaon pictus and cheetahs – Acinonyx jubatus) that defy apex and me- predators pursue prey (see Schmitz, 2003, 2008; Preisser et al., 2005). sopredator definitions given that they are technically mid-ranking In carnivore-ungulate systems, these three categories have typically carnivores in the systems in which they reside and yet, have predation been presented as two hunting modes represented by active and ambush risk effects that would best be described as persistent (FitzGibbon, strategies (see Hopcraft et al., 2005; Thaker et al., 2011; Middleton 1993; Ford and Goheen, 2015). For the purposes of this analysis, we et al., 2013; Moll et al., 2016; Petrunenko et al., 2016). Active carni- considered these species to be apex predators. vores are coursing predators that are regularly moving in pursuit of prey whereas ambush carnivores (including both sit-and-wait and sit- 2.3. Trophic level complexity and-pursue styles) are those that wait in a location and attack when a prey comes within a conventional chase distance. Thus, we designated Next, we documented the potential trophic complexity in Carnivora- each carnivore species as either an active of ambush predator. Ungulata interactions among the top three most-studied systems fea- Though not yet widely assessed in the literature, we also anticipate tured in our review. To do so, we developed lists of the carnivore and that the nature and strength of risk effects experienced by ungulate prey ungulate species commonly resident in these systems. Importantly should vary depending on whether the carnivore species functions as an however, we did not consider all interactions of carnivores and un- apex or mesopredator in the system. Here we use apex predator to mean gulates to be capable of eliciting risk effects. Due to body size, hunting the top-ranking carnivore(s) in the system, whereas mesopredators are mode, and sociality, there are many Carnivora-Ungulata interactions species that are mid-ranking (see Prugh et al., 2009). We consider the that are inherently non-predatory. Take for instance, red foxes (Vulpes risk effects from apex predators to be persistent while we envision the vulpes) that pose no predation risk to moose (Alces alces). In this case, risk effects of mesopredators to be better described as intermittent we would not consider red fox as part of the carnivore assemblage (Fig. 1). Though mesopredators can undoubtedly elicit risk effects in capable of eliciting risk effects in moose. Finally, we did not consider ungulates (Lingle, 2002; Bastille-Rousseau et al., 2015), their effects prey preference within this assessment of trophic level complexity. tend not to be consistent year round. Direct predation of ungulate prey by mesopredators is often restricted to neonates (Paquet, 1992; Linnell 3. Results et al., 1995; Arjo et al., 2002; Berger et al., 2008), creating an annual pulse in risk effects. Via the processes of mesopredator release, carni- Application of our search terms in Web of Science returned a total of vore species that are traditionally mesopredators can ascend to apex 339 studies. Following examination of these studies, we retained 170 predator positions. Coyotes (Canis latrans) provide a classic example of for analysis. These 170 studies directly assessed risk effects in carni- this premise (Ripple et al., 2013), because they function as mesopre- vore-ungulate systems (see Supplementary Table S1). We identified 82 dators in systems with larger carnivores, such as gray wolves (Canis carnivore-ungulate study systems in 25 countries across five continents. lupus; Berger et al., 2008), and as apex predators in systems where We found that 61% (n = 104 of 170) of these studies were centered in larger carnivores have been extirpated (Crooks and Soulé, 1999; North America, 19% (n = 33 of 170) in Africa, and 13% (n = 21 of 170) in Europe. A minority of these studies were situated in South America (5%, n = 8 of 170) and Asia (2%, n = 4 of 170). The majority (63%, n = 107 of 170) of these studies were single predator – single prey research designs while only one system (Isle Royale National Park, Michigan) among these studies that could be described as a single predator – single prey system (i.e., gray wolf – moose). Just 6% (n =11 of 170) were multiple predator – multiple prey designs. Over 82% (n = 141 of 170) of the studies featured one carnivore species and 74% (n = 125 of 170) of the studies featured one ungulate species. There were 34 studies (20%) that were single predator - multiple prey designs and 18 studies (11%) that were multiple predator - single prey designs. There were 29 studies (~17%) that assessed ≥2 carnivore species while 45 studies (26%) assessed ≥2 ungulate species. There were 11 studies (~6%) that assessed ≥3 carnivore species while 20 studies (12%) assessed ≥3 ungulate species. A total of 24 species of carnivores (Table 1) and 56 species of un- gulates (Table 2) were assessed in this literature. There was an average of 1.26 (SD = 0.66, range = 1 to 5) carnivore species researched per study with carnivores being used a total of 214 times to assess ungulate risk effects among this literature. The species of carnivore that was most-commonly evaluated was the gray wolf, which occurred 96 times (45%; Table 1). The second-ranked species was the African lion (Pan- thera leo; 11%) followed by coyotes (7%), and cougars (Puma concolor; 7%; Table 1). Most of the time (84%, n = 180 of 214) these species inhabited the apex predator position followed by mesopredators (8%, Fig. 1. A depiction of the trophic complexity in the order Carnivora and infra- n = 18 of 214). Coyotes were the species that could be either apex or order Ungulata in Hiawatha National Forest, Michigan, USA including apex mesopredator depending on the system and they were considered 7% predators, mesopredators, and ungulate prey. The dashed black lines represent (n = 16 of 214) of the time. Similarly, the majority (75%, n = 161 of potentially persistent risk effects deriving from the apex predators whereas the 214) of the time, carnivores among this literature had an active hunting dotted gray lines illustrate risk effects that could be described as being more mode with 25% (n = 53 of 214) having an ambush hunting mode. intermittent (i.e., seasonal) in nature. There was an average of 1.60 (SD = 1.42, range = 1 to 11) ungulate
3 R.A. Montgomery, et al. Biological Conservation 233 (2019) 1–11
Table 1 Table 2 The carnivore species, rank, hunting mode, and the number of times that each The ungulate species and the number of times that each ungulate was con- carnivore was considered among studies of risk effects in carnivore-ungulate sidered among studies of risk effects in carnivore-ungulate systems published systems studies published between 1989 and 2018. between 1989 and 2018.
Carnivore species Rank Hunting Count Proportion Ungulate species Count Proportion mode Cervus elaphus 66 0.24 Canis lupus Apex predator Active 96 0.45 Rangifer tarandus 21 0.08 Panthera leo Apex predator Ambush 23 0.11 Alces alces 15 0.06 Canis latrans Mesopredator or Active 16 0.07 Equus quagga 15 0.06 apex predator Connochaetes taurinus 14 0.05 Puma concolor Apex predator Ambush 15 0.07 Capreolus capreolus 13 0.05 Ursus arctos Apex predator Active 10 0.05 Odocoileus virginianus 13 0.05 Lycaon pictus Apex predator Active 9 0.04 Aepyceros melampus 9 0.03 Ursus americanus Apex predator Active 8 0.04 Tragelaphus strepsiceros 9 0.03 Lynx lynx Mesopredator Ambush 7 0.03 Odocoileus hemionus 8 0.03 Acinonyx jubatus Apex predator Active 6 0.03 Bison bison 6 0.02 Crocuta crocuta Apex predator Active 6 0.03 Phacochoerus africanus 6 0.02 Vulpes vulpes Mesopredator Active 3 0.01 Sus scrofa 6 0.02 Panthera leo persica Apex predator Ambush 2 0.009 Syncerus caffer 5 0.02 Panthera pardus Apex predator Ambush 2 0.009 Alcelaphus buselaphus 4 0.01 Canis anthus Apex predator Active 1 0.005 Bison bonasus 4 0.01 Canis lupus signatus Apex predator Active 1 0.005 Eudorcas thomsonii 4 0.01 Canis mesomelas Mesopredator Active 1 0.005 Giraffa camelopardalis 4 0.01 Lycalopex culpaeus Mesopredator Active 1 0.005 Lama guanicoe 4 0.01 Lynx canadensis Mesopredator Ambush 1 0.005 Equus burchelli 3 0.01 Lynx rufus Mesopredator Ambush 1 0.005 Nanger granti 3 0.01 Neofelis diardi Mesopredator Ambush 1 0.005 Axis axis 2 0.01 Panthera onca Apex predator Ambush 1 0.005 Hippotragus niger 2 0.01 Urocyon cinereoargenteus Mesopredator Active 1 0.005 Ourebia ourebi 2 0.01 Vulpes ferrilata Mesopredator Active 1 0.005 Tragelaphus oryx 2 0.01 Vulpes rueppellii Mesopredator Active 1 0.005 Vicugna vicugna 2 0.01 Antilocapra americana 1 0.004 Apennine chamois 1 0.004 species researched per study with ungulates being used a total of 272 Dama dama 1 0.004 ff Damaliscus korrigum 1 0.004 times to assess risk e ects among this literature. The species of ungulate Gazella dorcas 1 0.004 that was most-commonly evaluated was elk/red deer (Cervus elaphus)as Hippocamelus bisulcus 1 0.004 it was featured 66 times (24%; Table 2). The second-ranked species was Hippotragus equinus 1 0.004 caribou/reindeer (Rangifer tarandus; 8%), followed by moose (6%), and Kobus ellipsiprymnus 1 0.004 plains zebra (Equus quagga; 6%; Table 2). Among the 107 studies that Madoqua guentheri 1 0.004 Mazama americana 1 0.004 were single predator - single prey, the most commonly researched in- Mazama gouazoubira 1 0.004 teraction was gray wolf – elk/red deer occurring in 46% (n = 49 of 107) Muntiacus atherodes 1 0.004 of these studies (Table 3). The next most common interaction was Muntiacus muntjak 1 0.004 coyote – white-tailed deer (Odocoileus virginianus; 8%, n = 9 of 107), Oryx dammah 1 0.004 – Oryx gazella 1 0.004 followed by gray wolf caribou/reindeer (7%, n = 7 of 107), and gray Ovis canadensis mexicana 1 0.004 wolf – moose (6%, n = 6 of 107; Table 3). Ovis canadensis sierrae 1 0.004 Evaluated in 23% (n = 39 of 170) of the studies, the Greater Ovis dalli dalli 1 0.004 Yellowstone Ecosystem in the United States was the most-commonly Ovis dalli stonei 1 0.004 researched system overall. The next most-studied system was Banff Pecari tajacu 1 0.004 Phacochoerus aethiopicus 1 0.004 National Park, Canada (5%, n = 9 of 170), followed by Hwange Procapra przewalskii 1 0.004 National Park, Zimbabwe (4.7%, n = 8 of 170). In the Greater Raphicerus campestris 1 0.004 Yellowstone Ecosystem, there are eight ungulate species that are com- Rusa unicolor 1 0.004 monly resident and potentially experience predation risk from four Sus barbatus 1 0.004 Sylvicapra grimmia 1 0.004 resident apex predators and four resident mesopredators (Fig. 2a). Taurotragus oryx 1 0.004 Despite this complexity, the average number of carnivore species con- Tragelaphus buxtoni 1 0.004 sidered per study was 1.10 (SD = 0.38, range 1 to 3) and the average Tragulus kanchil 1 0.004 number of ungulate species considered per study was 1.15 (SD = 0.54, Tragulus napu 1 0.004 range 1 to 4; Fig. 2b). Banff National Park now has seven resident ungulate species (with bison – Bison bison being restored in 2017) that 4. Discussion potentially experience predation risk from four apex predators and three mesopredators (Fig. 3a). However, the only trophic interaction in Broadly, we detected a tendency to simplify complex systems among Banff National Park considered among these studies was gray wolf – elk carnivore-ungulate risk e ffects research. The majority of the studies (Fig. 3b). In Hwange National Park, there are eleven ungulate species reviewed (63%; n = 107 of 170) involved single predator - single prey that are commonly resident and potentially experience predation risk research designs, despite the fact that all but one system (e.g., Isle from five resident apex predators and two resident mesopredators Royale National Park, Michigan) had multiple resident carnivore and/ (Fig. 4a). Among these studies, there was just one carnivore species or ungulate species. It is a rare system where a prey species experiences considered per study and the average number of ungulate species predation risk from just one predator (see Sih et al., 1998; Lima, 2002; considered per study was 2.88 (SD = 3.36, range 1 to 11; Fig. 4b). Vanak et al., 2013). Thus, the tendency to simplify often contrasted with the complexity inherent to the systems in which these studies occurred. For example, the three most-studied systems in our review
4 R.A. Montgomery, et al. Biological Conservation 233 (2019) 1–11
Table 3 of studies focused on a single species pair, there are some notable ex- Trophic interactions among single predator – single prey studies of risk effects amples of carnivore-ungulate studies that did examine multispecies in carnivore-ungulate systems studies published between 1989 and 2018. interactions (see Valeix et al., 2009a, b; Thaker et al., 2011; Moll et al., Trophic interaction Count Proportion 2016; Creel et al., 2017; Dröge et al., 2017). The complexity that we discuss within this context should be in- Canis lupus - Cervus elaphus 49 0.46 terpreted as the maximum complexity in a system. We acknowledge Canis latrans - Odocoileus virginianus 9 0.08 that the strength of risk effects will importantly depend upon the rates Canis lupus - Rangifer tarandus 7 0.07 Canis lupus - Alces alces 6 0.06 at which predators and prey encounter one another (i.e., Holling, 1959; Lynx lynx - Capreolus capreolus 5 0.05 Lima and Dill, 1990; Middleton et al., 2013). These encounter rates Puma concolor - Odocoileus hemionus 4 0.04 vary according to life history characteristics, movement ecology, ha- Panthera leo - Equus quagga 3 0.03 bitat selection, and population dynamics of the predators and prey Puma concolor - Lama guanicoe 2 0.02 Panthera leo persica - Axis axis 2 0.02 (Hebblewhite et al., 2005a, b; Nilsen et al., 2009; Valeix et al., 2010; Ursus americanus - Rangifer tarandus 2 0.02 Montgomery et al., 2013, 2014). Thus, local species rarity might make Canis lupus - Bison bison 2 0.02 risk effects deriving from a certain carnivore negligible. In Banff Na- Puma concolor - Hippocamelus bisulcus 1 0.01 tional Park, Canada, for example, predation risk experienced by elk ff Panthera leo - Syncerus ca er 1 0.01 predominantly derives from gray wolves (Hebblewhite et al., 2005a, b; Crocuta crocuta - Tragelaphus buxtoni 1 0.01 Canis mesomelas - Gazella thomsonii 1 0.01 Hebblewhite and Merrill, 2007). Thus, predation risk from the sympa- Lycaon pictus - Madoqua guentheri 1 0.01 tric, but comparatively rare, grizzly bear (Ursus arctos) might not induce Puma concolor - Rangifer tarandus 1 0.01 strong effects. However, we have found that research in carnivore-un- Canis lupus - Sus scrofa 1 0.01 gulate systems to date has typically not been complex enough to Canis lupus - Apennine chamois 1 0.01 ff Puma concolor - Mazama americana 1 0.01 quantify the strength of risk e ects deriving from multiple species of Ursus arctos - Alces alces 1 0.01 carnivore. Instead, this research has tended to focus on more obvious or Puma concolor - Vicugna vicugna 1 0.01 dominant elements of the trophic system (i.e., gray wolves and elk/red Canis lupus - Bison bonasus 1 0.01 deer). Panthera leo - Connochaetes taurinus 1 0.01 We also acknowledge that trophic complexity will continue to Ursus arctos - Rangifer tarandus 1 0.01 Panthera leo - Aepyceros melampus 1 0.01 change according to natural (e.g., immigration and emigration) or an- Panthera leo - Alcelaphus buselaphus 1 0.01 thropogenic conditions. For instance, anthropogenic changes are on- going in Banff National Park where bison were restored starting in 2017 (Steenweg et al., 2016). In Europe, recolonization of larger carnivores (Greater Yellowstone Ecosystem, Banff National Park, and Hwange to human-dominated landscapes is a very dynamic process (Chapron National Park) have a minimum of five carnivore species and seven et al., 2014). Reintroductions of large carnivores to systems in which ungulate species per system (Figs. 2a, 3a, and 4a). Nevertheless, a small they once occurred is increasingly common, particularly among pro- fraction of this complexity tended to be assessed among the studies in tected areas in Africa (Hayward et al., 2007; Davies et al., 2016; Makin this review (Figs. 2b, 3b, and 4b). In interpreting these results, we et al., 2017). It was beyond the scope of our assessment to consider the would first like to acknowledge how challenging it can prove to be to ways in which trophic complexity might vary with space or time. study multispecies interactions and risk effects in carnivore-ungulate The studies in our assessment also tended to consider the risk effects systems. Though carnivore-ungulate systems are immensely valuable deriving from carnivores with an active hunting mode. Species with an for the study of risk effects, developing experiments that can elucidate ambush style hunting mode were evaluated just 25% of the time the nature and strength of those effects is non-trivial (see Creel et al., (n = 53 of 214). This result is in keeping with research that shows that 2017; Peers et al., 2018). The simultaneous assessment of numerous ambush hunters, which are generally less conspicuous than active species of ungulates under threat from several sympatric carnivore hunters, are three times less likely to be the focus of research than species presents considerable experimental, financial, and logistic active hunters despite the often times higher densities of ambush pre- constraints. Despite these considerations, and the fact that the majority dators in ecological systems. Despite these trends, the strongest risk
Fig. 2. The trophic complexity of potential carnivore-ungulate interactions in the Greater Yellowstone Ecosystem, United States (panel a) and the actual trophic interactions studied (panel b) among the literature assessing risk effects in carnivore-ungulate systems between 1989 and 2018. This figure includes the common ungulate species in the park. The dashed black line depicts persistent risk effects deriving from the apex predator(s) while the gray circle line depicts intermittent risk effects deriving from the mesopredator.
5 R.A. Montgomery, et al. Biological Conservation 233 (2019) 1–11
Fig. 3. The trophic complexity of potential carnivore-ungulate interactions in Banff National Park, Canada (panel a) and the actual trophic interactions studied (panel b) among the literature assessing risk effects in carnivore-ungulate systems between 1989 and 2018. This figure includes the common ungulate species in the park. However, bison (Bison bison) were only restored to the park in 2017. The dashed black line depicts persistent risk effects deriving from the apex predator. effects are predicted to derive from ambush (i.e., sit-and-wait and sit- habitat domains of predators and prey might help predict whether and-pursue) predators given that their cues are indicative of imminent density-mediated or trait-mediated indirect effects predominate. Thus, predation risk (Lima and Bednekoff, 1999; Schmitz, 2005; Preisser the dynamics of multiple predators and prey are suggested to be able to et al., 2007; Miller et al., 2014; Schmitz et al., 2017). In contrast, cues be predicted across the spatial domain of their interactions to quantify from active predators are typically diffused by the almost constant the mechanisms associated with trophic functioning (McCann et al., movement (Preisser et al., 2007; Miller et al., 2014; Schmitz et al., 2005; Barraquand and Murrell, 2013; Northfield et al., 2017). This 2017). Thus, it is possible that ambush carnivores are inducing even demonstrates the potential power of studying multispecies interactions greater risk effects in ungulate prey than are active carnivores. For in carnivore-ungulate systems. However, while this framework has example, the habitat domain conceptual theory suggests that the nature been suggested to be generalizable to large vertebrate communities (see and strength of consumptive and non-consumptive effects can be pre- Schmitz et al., 2017) it has yet to be rigorously applied to these systems. dicted as a function of predator hunting mode and habitat domain We see this as a potential area of growth in carnivore-ungulate re- (Schmitz, 2008; Miller et al., 2014; Northfield et al., 2017). Here, ha- search. bitat domain refers to the extent of microhabitat used by an animal in We also acknowledge that risk effects might vary according to support of their foraging where predator-prey interactions can occur predator rank. Specifically, we presented the risk effects of apex pre- (Schmitz et al., 2004; Preisser et al., 2007; Schmitz et al., 2017). dators as persistent (i.e., relevant year-round) and the risk effects of Coarsely, animals can be broad or narrow habitat domain organisms mesopredators to be intermittent (i.e., subject to seasonal variation; (Schmitz et al., 2004). Hunting mode determines the width of the Fig. 1). The current ability to attribute risk effects to carnivores with predator habitat domain, where ambush predators have narrow habitat specific ranks and hunting modes is limited (sensu Atwood et al., 2009) domains and active predators have broad habitat domains (Schmitz given that only 17% of the studies considered multiple predators. Thus, et al., 2017). Prey habitat domain size is determined by life history we recommend that researchers work to delineate and quantify the traits including dietary requirements and foraging mode (Northfield nature and strength of risk effects deriving from both predator rank and et al., 2017; Schmitz et al., 2017). The amount of overlap among the predator hunting mode. Importantly however, additional research is
Fig. 4. The trophic complexity of potential carnivore-ungulate interactions in Hwange National Park, Zimbabwe (panel a) and the actual trophic interactions studied (panel b) among the literature assessing risk effects in carnivore-ungulate systems between 1989 and 2018. This figure includes the common ungulate species in the park. The dashed black line depicts persistent risk effects deriving from the apex predator(s).
6 R.A. Montgomery, et al. Biological Conservation 233 (2019) 1–11 needed to determine whether the interactions of multiple carnivores is Though these models have been instrumental in our collective under- best characterized by exploitative competition (e.g., sympatric pre- standing of predator-prey ecology, they have been widely criticized for dators consume the same prey), interference competition (e.g., the being oversimplified (Ayala et al., 1973; Tilman, 1987; Berryman, territoriality of sympatric predators excludes one another from pur- 1992). In addition to being overly simple in terms of the number of suing prey), or intraguild predation (e.g., sympatric predators prey species considered, they also tend to lack a spatial dimension. Spatial upon each other; see Gotelli, 2008; Schmitz et al., 2017). These di- variation is, of course, integral to risk effects research (Heithaus and mensions have important implications for niche complementarity Dill, 2002; Mitchell and Lima, 2002; Wirsing et al., 2007; Laundré, among sympatric carnivores (sensu Schoener, 1974). Similarly, the in- 2010). Risk effects depend on both the amount of resources and refugia teractions of multiple sympatric ungulate species could modulate the in the system and the tradeoffs that prey make herein (e.g. McNamara risk effects experienced by any one prey species. It is widely appre- and Houston, 1987; Peacor, 2003; Schmitz et al., 2004). Prey have to ciated, particularly on the African savanna, that inter-species grouping choose between foraging in a situation where risk of predation is high, can be an effective antipredator behavior (Fryxell, 1995; Krause et al., to abandon foraging, or to forage in habitat where resource quality is 2002; Thaker et al., 2011; Schmitt et al., 2016). Not without its tra- lower so as to reduce predation risk (Abrams, 1984; Brown, 1999; Lima deoffs, species can co-mingle as a means of dispersing shared vigilance and Dill, 1990; Peacor, 2003; Schmitz et al., 2004). across the group. However, here again, there are important connections The interactions of carnivores and ungulates are complex and de- with predator hunting mode given that group size can alter the sus- pend on a diversity of factors including characteristics of: i) the prey ceptibility of prey to either active or ambush predators (Parrish, 1993; (Creel, 2011), ii) the environment (Acebes et al., 2013; Riginos, 2015), Scheel, 1993; Pays et al., 2007). This emphasizes the need for addi- and/or iii) the predator (e.g. hunting mode; see Creel et al., 2014; tional research to quantify the consequences of interspecific anti- Schmitz et al., 2017). These factors can affect the rates at which species predator decisions on risk effects. encounter one another in space and time (Middleton et al., 2013). The tendency to simplify complex carnivore-ungulate systems that Further, empirical evidence on the simultaneous and dynamic nature of we observed is striking, but not greatly surprising. Here we discuss the the movement tactics of both predators and prey in large vertebrate ways in which simplicity is a cornerstone value in science, predator- systems has accumulated with recent demonstration of large-scale prey theory has largely been developed via examinations of single flights (a few kilometres) of zebras to avoid the risk of predation by predator – single prey interactions, and risk effects research was African lions both proactively (Courbin et al., 2019) and reactively founded on examinations of small organisms in small systems. (Courbin et al., 2016), in parallel to an active rotation of hunting grounds by lions at the landscape scale (Valeix et al., 2011). Thus, 4.1. The value of simplicity in science spatio-temporal variation should be integral to models explaining risk effects in carnivore-ungulate systems. For example, circumstances that Simplicity has been an enduring value in science for centuries affect the local abundance or occurrence of one prey species, might lead (Forster and Sober, 1994). This value is often encapsulated by Occam's to prey switching among a generalist carnivore (e.g. Patterson et al., Razor, which postulates that when competing hypotheses explain a 1998; Garrott et al., 2007). Research demonstrates that the functional phenomenon equally well, the simplest among them is to be preferred response of predators changes in line with temporal variation in prey (Sober, 2015). Occam's Razor is variously restated and often referred to abundance (e.g. Höner et al., 2002). Furthermore, spatial variation can as the principle of parsimony in ecological literature (Tukey, 1961; Box lead directly and indirectly impact inter-species interactions. In the and Jenkins, 1970; Burnham and Anderson, 2002). Widely accepted in Greater Yellowstone Ecosystem, elk have been found to shift to habitats ecology (e.g., Kimmins et al., 2008; Engström et al., 2016) as having that have more complex structure to reduce predation risk from wolves value in building reliable knowledge (cf. Evans et al., 2013), the central (active predators) which increases their predation risk from cougars challenge of Occam's Razor relates to the determination that competing (sit-and-wait predators; Atwood et al., 2009). Within this context, as- hypotheses explain the data “equally well” (Baker, 2007). In statistical sessing only the gray wolf – elk interaction would obscure the im- modeling, such decisions are often carried out via information theoretic portant indirect effects associated with cougars. approaches that compare predictive accuracy with model complexity The simplistic origins of predator-prey theory are likely one of the (Akaike, 1974; Burnham and Anderson, 2002). In broader practice reasons why Isle Royale National Park has become such a celebrated however, there is often considerable debate regarding contexts in which and textbook example of carnivore-ungulate interactions. Isle Royale more complex models are preferable to their simpler counterparts (see was the only system in our review that could be described as a single Merow et al., 2014). predator – single prey system (Peterson and Page, 1988; Montgomery The complexity inherent to carnivore-ungulate interactions, for et al., 2013, 2014). Importantly, Isle Royale is an island system with a example, is difficult to capture. A key challenge then becomes ap- fairly unique carnivore – ungulate history. For instance, the fact that the proximating (i.e., the ‘art of approximation’) this complexity well en- system is an island mean that immigration and emigration of the two ough to generate useful inference (Akaike, 1974; Burnham and species (gray wolves and moose) is negligible (Adams et al., 2011). Anderson, 2002; Holt and Slade, 2004). One heuristic strategy is to Consequently, the gray wolves on the island experienced high rates of begin with simple approaches or models and increase their complexity inbreeding which, in addition to a number of other factors, led to the only after it is demonstrated that they fail to explain or predict phe- demise of the population (Räikkönen et al., 2009; Hedrick et al., 2014, nomena well (Rosindell et al., 2012). Here we have documented that 2016). Via a series of reintroductions, gray wolves are now being re- the first ~25 years of risk effects in carnivore-ungulate systems has stored to the island and so the single predator – single prey nature of been relatively simple. Thus, now may be the launching point to de- this island ecosystem will continue (see Mlot, 2018). However, the velop more sophisticated and multi-species approaches building on top unique simplicity of carnivore-ungulate trophic interactions on Isle of these comparatively simple designs. Royale, does not necessarily translate to other systems with broader guilds of sympatric carnivores and ungulates. 4.2. Single species dynamics in predator prey theory 4.3. Small species origins of risk effects research It is important to note that the inception of predator prey theory involved deterministic and often reductive measures (Chesson, 1978). Risk effects research, in particular, has been established via the Take for example, the differential equations of Lotka (1925), Volterra study of predator-prey interactions of relatively small species inhabiting (1926), and Nicholson and Bailey (1935) which were all originally fit relatively small systems (Schmitz et al., 2017). The intent of risk effects using the input of data from a single predator and single prey species. research is often to determine if antipredator behaviors have fitness-
7 R.A. Montgomery, et al. Biological Conservation 233 (2019) 1–11 based consequences (Creel and Christianson, 2008; Preisser et al., 2005; examining multispecies effects in complex carnivore-ungulate systems Schmitz et al., 1997). These nonconsumptive effects are non-trivial to to quantify the ways in which these interactions may provide unique assess and often require experimental and longitudinal components to insights into the functioning of these systems. As risk effects may scale the study design. For example, it can take only a few days to implement to have population-level consequences, this research can be expected to an experiment assessing the effect of predation risk on the growth rate be directly applicable to prevailing conservation practice and should of an aphid population (Nelson et al., 2004), whereas 10 years were inform the implementation of progressive and effective policies de- needed to assess the risk effects of a mammalian carnivore on the signed to protect these species. snowshoe hare (Lepus americanus; Boonstra et al., 1998). Furthermore, experimental designs that manipulate the lethal ability of predators Appendix A. Supplementary data (e.g. Peckarsky et al., 1993; Schmitz et al., 1997; Werner and Peacor, 2006) have been commonly deployed to disentangle the role of lethal Supplementary data to this article can be found online at https:// and risk effects. These techniques are neither practical nor ethical in the doi.org/10.1016/j.biocon.2019.02.001. case of carnivores hunting ungulates in vast terrestrial systems. Thus, experimentation remains an enduring challenge of risk effects research References in carnivore-ungulate systems (Sih et al., 1998; Schmitz et al., 2004; Creel et al., 2008; Creel, 2011; Gehr et al., 2018; Peers et al., 2018). Abrams, P.A., 1984. Foraging time optimization and interactions in food webs. Am. Nat. Furthermore, much of the risk effects research in carnivore-ungulate 124 (1), 80–96. Abrams, P.A., 1995. Implications of dynamically variable traits for identifying, classi- systems has been positioned in North America and Africa. Though there fying, and measuring direct and indirect effects in ecological communities. Am. Nat. is tremendous spatial variation, these two continents continue to 146 (1), 112–134. maintain comparatively high levels of carnivore-ungulate species di- Acebes, P., Malo, J.E., Traba, J., 2013. Trade-offs between food availability and predation risk in desert environments: the case of polygynous monomorphic guanaco (Lama versity (Ripple et al., 2014, 2016). We acknowledge that South America guanicoe). J. Arid Environ. 97, 136–142. and Australia were underrepresented in our analysis given that our Adams, J.R., Vucetich, L.M., Hedrick, P.W., Peterson, R.O., Vucetich, J.A., 2011. Genomic interest was to specifically assess the risk effects that might derive from sweep and potential genetic rescue during limiting environmental conditions in an – the interactions of carnivores and their ungulate prey. Undoubtedly, isolated wolf population. Proc. R. Soc. B 278 (1723), 3336 3344. Akaike, H., 1974. A new look at the statistical model identification. IEEE Trans. Autom. there are many systems around the world where the primary prey of Control 19 (6), 716–723. large carnivores are not ungulate species. Take for example jaguars Arjo, W.M., Pletscher, D.H., Ream, R.R., 2002. Dietary overlap between wolves and – hunting capybara (Hydrochoerus hydrochaeris) in South America coyotes in northwestern Montana. J. Mammal. 83 (3), 754 766. Atwood, T.C., Gese, E.M., Kunkel, K.E., 2009. Spatial partitioning of predation risk in a (Schaller and Vasconcelos, 1978) or predation of red kangaroos (Mac- multiple predator-multiple prey system. J. Wildl. Manag. 73 (6), 876–884. ropus rufus) by dingos in Australia (Corbett and Newsome, 1987). These Ayala, F.J., Gilpin, M.E., Ehrenfeld, J.G., 1973. Competition between species: theoretical exclusions in no way diminish the importance of these interactions in models and experimental tests. Theor. Popul. Biol. 4 (3), 331–356. ff Baker, A., 2007. Occam's razor in science: a case study from biogeography. Biol. Philos. risk e ects research, but rather that they were not integral to our as- 22 (2), 193–215. sessment of carnivore-ungulate research. Barnes, C., Maxwell, D., Reuman, D.C., Jennings, S., 2010. Global patterns in pre- dator–prey size relationships reveal size dependency of trophic transfer efficiency. Ecology 91 (1), 222–232. 5. Implications for conservation Barraquand, F., Murrell, D.J., 2013. Scaling up predator–prey dynamics using spatial moment equations. Methods Ecol. Evol. 4 (3), 276–289. The simplicity inherent to the risk effects research in carnivore- Bastille-Rousseau, G., Potts, J.R., Schaefer, J.A., Lewis, M.A., Ellington, E.H., Rayl, N.D., Mahoney, S.P., Murray, D.L., 2015. Unveiling trade-offs in resource selection of mi- ungulate systems raises two primary concerns for conservation. First, gratory caribou using a mechanistic movement model of availability. Ecography 38 via the omission of one or more species, the oversimplification of these (10), 1049–1059. systems might generate results that are misleading or too general to be Berger, K.M., Gese, E.M., Berger, J., 2008. Indirect effects and traditional trophic cas- – particularly useful. By excluding species that might contribute to- or cades: a test involving wolves, coyotes, and pronghorn. Ecology 89 (3), 818 828. Bernot, R.J., Turner, A.M., 2001. Predator identity and trait-mediated indirect effects in a modulate risk effects, important dynamics driven by the omitted species littoral food web. Oecologia 129 (1), 139–146. are being missed. This means that we don't yet have a full under- Berryman, A.A., 1992. The Orgins and evolution of predator-prey theory. Ecology 73 (5), – standing of the ways in which risk effects impact carnivore-ungulate 1530 1535. Boonstra, R., Hik, D., Singleton, G.R., Tinnikov, A., 1998. The impact of predator-induced systems. The second important issue regarding simplicity relates to the stress on the snowshoe hare cycle. Ecol. Monogr. 68 (3), 371–394. generalization of results from a specific carnivore-ungulate trophic in- Box, G., Jenkins, G.M., 1970. Time Series Analysis: Forecasting and Control. Holden-Day, teraction to carnivore-ungulate systems more broadly. For example, it is London. Brose, U., Jonsson, T., Berlow, E.L., Warren, P., Banasek-Richter, C., Bersier, L.F., doubtful whether the behavioral risk dynamics of wolves and elk in Blanchard, J.L., Brey, T., Carpenter, S.R., Blandenier, M.F.C., Cushing, L., 2006. Yellowstone National Park can be extrapolated to other wolf-ungulate Consumer–resource body-size relationships in natural food webs. Ecology 87 (10), systems in other landscapes (Schmidt and Kuijper, 2015), let alone 2411–2417. ff Brown, J.S., 1999. Vigilance, patch use and habitat selection: foraging under predation systems with di erent species of large carnivores (Moll et al., 2017). risk. Evol. Ecol. Res. 1 (1), 49–71. Via the simplification of complex systems we have the potential to Brown, J.S., Kotler, B.P., 2004. Hazardous duty pay and the foraging cost of predation. misunderstand and misrepresent the mechanisms that govern carni- Ecol. Lett. 7 (10), 999–1014. Brown, J.S., Kotler, B.P., Smith, R.J., Wirtz, W.O., 1988. The effects of owl predation on vore-prey interactions. This is particularly concerning given the po- the foraging behavior of heteromyid rodents. Oecologia 76 (3), 408–415. tential fitness ramifications of risk effects. In combination with lethal Brown, J.S., Laundré, J.W., Gurung, M., 1999. The ecology of fear: optimal foraging, effects, the nature and strength of risk effects alter the population dy- game theory, and trophic interactions. J. Mammal. 80 (2), 385–399. namics of prey with subsequent implications for predator populations. Burnham, K.P., Anderson, D.R., 2002. Model Selection and Multimodel Inference: A Practical Information-Theoretic Approach, 2nd ed. Springer. Thus, failure to appropriately comprehend the interspecies interactions Chapron, G., Kaczensky, P., Linnell, J.D., von Arx, M., Huber, D., Andrén, H., López-Bao, that lead to these population-level consequences can lead to the for- J.V., Adamec, M., Álvares, F., Anders, O., Balčiauskas, L., 2014. Recovery of large mation of inappropriate policies meant to be conserving these carni- carnivores in Europe's modern human-dominated landscapes. Science 346 (6216), 1517–1519. vore-ungulate systems. This is particularly concerning within the con- Chesson, P., 1978. Predator-prey theory and variability. Annu. Rev. Ecol. Syst. 9 (1), text of carnivore-ungulate systems given that over three-quarters of the 323–347. 31 species of large carnivores remaining on the planet have populations Cohen, J.E., Newman, C.M., 1985. A stochastic theory of community food webs I. Models and aggregated data. Proc. R. Soc. Lond. B 224 (1237), 421–448. that are declining (Gittleman et al., 2001; Ripple et al., 2014). Fur- Cohen, J.E., Jonsson, T., Carpenter, S.R., 2003. Ecological community description using thermore, ungulate population declines present one of the greatest the food web, species abundance, and body size. Proc. Natl. Acad. Sci. 100 (4), conservation challenges of the 21st century (Ripple et al., 2016; Wolf 1781–1786. Corbett, L.K., Newsome, A.E., 1987. The feeding ecology of the dingo. Oecologia 74 (2), and Ripple, 2016). Thereby, we encourage researchers to consider
8 R.A. Montgomery, et al. Biological Conservation 233 (2019) 1–11
215–227. Bayley, S.E., Paquet, P.C., 2005a. Human activity mediates a trophic cascade caused Courbin, N., Loveridge, A.J., Macdonald, D.W., Fritz, H., Valeix, M., Makuwe, E.T., by wolves. Ecology 86 (8), 2135–2144. Chamaillé-Jammes, S., 2016. Reactive responses of zebras to lion encounters shape Hebblewhite, M., Merrill, E.H., McDonald, T.L., 2005b. Spatial decomposition of preda- their predator–prey space game at large scale. Oikos 125 (6), 829–838. tion risk using resource selection functions: an example in a wolf–elk predator–prey Courbin, N., Loveridge, A.J., Fritz, H., Macdonald, D.W., Patin, R., Valeix, M., Chamaillé- system. Oikos 111 (1), 101–111. Jammes, S., 2019. Zebra diel migrations reduce encounter risk with lions at night. J. Hedrick, P.W., Peterson, R.O., Vucetich, L.M., Adams, J.R., Vucetich, J.A., 2014. Genetic Anim. Ecol. 88, 92–101. rescue in Isle Royale wolves: genetic analysis and the collapse of the population. Creel, S., 2011. Toward a predictive theory of risk effects: hypotheses for prey attributes Conserv. Genet. 15 (5), 1111–1121. and compensatory mortality. Ecology 92 (12), 2190–2195. Hedrick, P.W., Kardos, M., Peterson, R.O., Vucetich, J.A., 2016. Genomic variation of Creel, S., 2018. The control of risk hypothesis: reactive vs. proactive antipredator re- inbreeding and ancestry in the remaining two Isle Royale wolves. J. Hered. 108 (2), sponses and stress-mediated vs. food-mediated costs of response. Ecol. Lett. 21 (7), 120–126. 947–956. Heithaus, M.R., Dill, L.M., 2002. Food availability and tiger shark predation risk influence Creel, S., Christianson, D., 2008. Relationships between direct predation and risk effects. bottlenose dolphin habitat use. Ecology 83 (2), 480–491. Trends Ecol. Evol. 23 (4), 194–201. Heithaus, M., Frid, A., Dill, L., 2002. Shark-inflicted injury frequencies, escape ability, Creel, S., Winnie, J.A., Christianson, D., Liley, S., 2008. Time and space in general models and habitat use of green and loggerhead turtles. Mar. Biol. 140 (2), 229–236. of antipredator response: tests with wolves and elk. Anim. Behav. 76 (4), 1139–1146. Heithaus, M.R., Frid, A., Wirsing, A.J., Dill, L.M., Fourqurean, J.W., Burkholder, D., Creel, S., Schuette, P., Christianson, D., 2014. Effects of predation risk on group size, Thomson, J., Bejder, L., 2007. State-dependent risk-taking by green sea turtles vigilance, and foraging behavior in an African ungulate community. Behav. Ecol. 25 mediates top-down effects of tiger shark intimidation in a marine ecosystem. J. Anim. (4), 773–784. Ecol. 76 (5), 837–844. Creel, S., Dröge, E., M'soka, J., Smit, D., Becker, M., Christianson, D., Schuette, P., 2017. Heithaus, M.R., Wirsing, A.J., Thomson, J.A., Burkholder, D.A., 2008. A review of lethal The relationship between direct predation and antipredator responses: a test with and non-lethal effects of predators on adult marine turtles. J. Exp. Mar. Biol. Ecol. multiple predators and multiple prey. Ecology 98 (8), 2081–2092. 356 (1–2), 43–51. Cresswell, W., 1993. Escape responses by redshanks, Tringa totanus, on attack by avian Holling, C.S., 1959. The components of predation as revealed by a study of small-mammal predators. Anim. Behav. 46 (3), 609–611. predation of the European pine sawfly. Can. Entomol. 91 (5), 293–320. Cresswell, W., 1994. Flocking is an effective anti-predation strategy in redshanks, Tringa Holt, R.D., Slade, N.A., 2004. Commentary. In: Taper, M.L., Lele, S.R. (Eds.), The Nature totanus. Anim. Behav. 47 (2), 433–442. of Scientific Evidence: Statistical, Philosophical, and Empirical Considerations. Cresswell, W., 2008. Non-lethal effects of predation in birds. Ibis 150 (1), 3–17. University of Chicago Press, Chicago, IL, USA. Cresswell, W., Quinn, J.L., 2013. Contrasting risks from different predators change the Höner, O.P., Wachter, B., East, M.L., Hofer, H., 2002. The response of spotted hyaenas to overall nonlethal effects of predation risk. Behav. Ecol. 24 (4), 871–876. long-term changes in prey populations: functional response and interspecific klep- Crooks, K.R., Soulé, M.E., 1999. Mesopredator release and avifaunal extinctions in a toparasitism. J. Anim. Ecol. 71 (2), 236–246. fragmented system. Nature 400 (6744), 563. Hopcraft, J.G.C., Sinclair, A.R.E., Packer, C., 2005. Planning for success: Serengeti lions D'amen, M., Rahbek, C., Zimmermann, N.E., Guisan, A., 2017. Spatial predictions at the seek prey accessibility rather than abundance. J. Anim. Ecol. 74 (3), 559–566. community level: from current approaches to future frameworks. Biol. Rev. 92 (1), Jonsson, T., Cohen, J.E., Carpenter, S.R., 2005. Food webs, body size, and species 169–187. abundance in ecological community description. Adv. Ecol. Res. 36, 1–84. Davies, A.B., Tambling, C.J., Kerley, G.I., Asner, G.P., 2016. Limited spatial response to Kerfoot, W.C., Sih, A., 1987. Predation: Direct and Indirect Impacts on Aquatic direct predation risk by African herbivores following predator reintroduction. Ecol. Communities. University Press of New England. Evol. 6 (16), 5728–5748. Kimmins, J.P.H., Blanco, J.A., Seely, B., Welham, C., Scoullar, K., 2008. Complexity in Dröge, E., Creel, S., Becker, M.S., M'soka, J., 2017. Risky times and risky places interact to modelling forest ecosystems: how much is enough? For. Ecol. Manag. 256 (10), affect prey behaviour. Nat. Ecol. Evol. 1 (8), 1123. 1646–1658. Elton, C.S., 1927. Animal Ecology. University of Chicago Press, Chicago, IL. Kotler, B.P., 1984. Risk of predation and the structure of desert rodent communities. Engström, K., Rounsevell, M.D., Murray-Rust, D., Hardacre, C., Alexander, P., Cui, X., Ecology 65 (3), 689–701. Palmer, P.I., Arneth, A., 2016. Applying Occam's razor to global agricultural land use Krause, J., Ruxton, G.D., Ruxton, G.D., 2002. Living in Groups. Oxford University Press. change. Environ. Model. Softw. 75, 212–229. LaManna, J.A., Martin, T.E., 2016. Costs of fear: behavioural and life-history responses to Estes, J.A., 1995. Top-level carnivores and ecosystem effects: questions and approaches. risk and their demographic consequences vary across species. Ecol. Lett. 19 (4), In: Linking Species & Ecosystems. Springer, Boston, MA, pp. 151–158. 403–413. Estes, J.A., Terborgh, J., Brashares, J.S., Power, M.E., Berger, J., Bond, W.J., Carpenter, Laundré, J.W., 2010. Behavioral response races, predator–prey shell games, ecology of S.R., Essington, T.E., Holt, R.D., Jackson, J.B., Marquis, R.J., 2011. Trophic down- fear, and patch use of pumas and their ungulate prey. Ecology 91 (10), 2995–3007. grading of planet Earth. Science 333 (6040), 301–306. Layman, C.A., Giery, S.T., Buhler, S., Rossi, R., Penland, T., Henson, M.N., Bogdanoff, Evans, M.R., Grimm, V., Johst, K., Knuuttila, T., De Langhe, R., Lessells, C.M., Merz, M., A.K., Cove, M.V., Irizarry, A.D., Schalk, C.M., Archer, S.K., 2015. A primer on the O'Malley, M.A., Orzack, S.H., Weisberg, M., Wilkinson, D.J., 2013. Do simple models history of food web ecology: fundamental contributions of fourteen researchers. Food lead to generality in ecology? Trends Ecol. Evol. 28 (10), 578–583. Webs 4, 14–24. Finke, D.L., Denno, R.F., 2004. Predator diversity dampens trophic cascades. Nature 429 Lima, S.L., 1998. Nonlethal effects in the ecology of predator-prey interactions. (6990), 407. Bioscience 48 (1), 25–34. FitzGibbon, C.D., 1993. Cheetahs and gazelles: a study of individual variation in anti- Lima, S.L., 2002. Putting predators back into behavioral predator–prey interactions. predator behaviour and predation risk. Physiol. Ecol. Jpn 29, 195–206. Trends Ecol. Evol. 17 (2), 70–75. Fleming, P.J., Nolan, H., Jackson, S.M., Ballard, G.A., Bengsen, A., Brown, W.Y., Meek, Lima, S.L., Bednekoff, P.A., 1999. Temporal variation in danger drives antipredator be- P.D., Mifsud, G., Pal, S.K., Sparkes, J., 2017. Roles for the Canidae in food webs havior: the predation risk allocation hypothesis. Am. Nat. 153 (6), 649–659. reviewed: where do they fit? Food Webs 12, 14–34. Lima, S.L., Dill, L.M., 1990. Behavioral decisions made under the risk of predation: a Ford, A.T., Goheen, J.R., 2015. An experimental study on risk effects in a dwarf antelope, review and prospectus. Can. J. Zool. 68 (4), 619–640. Madoqua guentheri. J. Mammal. 96 (5), 918–926. Lindeman, R.L., 1942. The trophic-dynamic aspect of ecology. Ecology 23 (4), 399–417. Forster, M., Sober, E., 1994. How to tell when simpler, more unified, or less ad hoc Lingle, S., 2002. Coyote predation and habitat segregation of white-tailed deer and mule theories will provide more accurate predictions. Br. J. Philos. Sci. 45 (1), 1–35. deer. Ecology 83 (7), 2037–2048. Fryxell, J.M., 1995. Aggregation and migration by grazing ungulates in relation to re- Linnell, J.D., Aanes, R., Andersen, R., 1995. Who killed Bambi? The role of predation in sources and predators. In: Sinclair, A.R.E., Arcese, P. (Eds.), Serengeti II: Dynamics, the neonatal mortality of temperate ungulates. Wildl. Biol. 1 (4), 209–223. Management, and Conservation of an Ecosystem. University of Chicago Press, Lotka, A.J., 1925. Elements of Physical Biology, by Alfred J. Lotka. Chicago, pp. 257–273. MacLeod, K.J., Krebs, C.J., Boonstra, R., Sheriff, M.J., 2018. Fear and lethality in snow- Garrott, R.A., Bruggeman, J.E., Becker, M.S., Kalinowski, S.T., White, P.J., 2007. shoe hares: the deadly effects of non-consumptive predation risk. Oikos 127 (3), Evaluating prey switching in wolf–ungulate systems. Ecol. Appl. 17 (6), 1588–1597. 375–380. Gehr, B., Hofer, E.J., Ryser, A., Vimercati, E., Vogt, K., Keller, L.F., 2018. Evidence for Makin, D.F., Chamaillé-Jammes, S., Shrader, A.M., 2017. Changes in feeding behavior nonconsumptive effects from a large predator in an ungulate prey? Behav. Ecol. 29 and patch use by herbivores in response to the introduction of a new predator. J. (3), 724–735. Mammal. 99 (2), 341–350. Gittleman, J.L., Funk, S.M., Macdonald, D., Wayne, R.K., 2001. Carnivore Conservation. Mangel, M., Clark, C.W., 1988. Dynamic Modeling in Behavioral Ecology. Princeton Cambridge University Press. University Press. Gotelli, N.J., 2008. A primer of ecology (No. 577.88 G6). Sinauer Associates, Sunderland, McCann, K.S., Rasmussen, J.B., Umbanhowar, J., 2005. The dynamics of spatially coupled Massachusetts, USA. food webs. Ecol. Lett. 8 (5), 513–523. Haidir, I.A., Macdonald, D.W., Linkie, M., 2018. Assessing the spatiotemporal interactions McCauley, D.J., Gellner, G., Martinez, N.D., Williams, R.J., Sandin, S.A., Micheli, F., of mesopredators in Sumatra's tropical rainforest. PLoS One 13 (9), e0202876. Mumby, P.J., McCann, K.S., 2018. On the prevalence and dynamics of inverted Hairston, N.G., Smith, F.E., Slobodkin, L.B., 1960. Community structure, population trophic pyramids and otherwise top-heavy communities. Ecol. Lett. 21 (3), 439–454. control, and competition. Am. Nat. 94 (879), 421–425. McNamara, J.M., Houston, A.I., 1987. Starvation and predation as factors limiting po- Hayward, M.W., Kerley, G.I., Adendorff, J., Moolman, L.C., O'Brien, J., Sholto-Douglas, pulation size. Ecology 68 (5), 1515–1519. A., Bissett, C., Bean, P., Fogarty, A., Howarth, D., Slater, R., 2007. The reintroduction Merow, C., Smith, M.J., Edwards Jr., T.C., Guisan, A., McMahon, S.M., Normand, S., of large carnivores to the Eastern Cape, South Africa: an assessment. Oryx 41 (2), Thuiller, W., Wüest, R.O., Zimmermann, N.E., Elith, J., 2014. What do we gain from 205–214. simplicity versus complexity in species distribution models? Ecography 37 (12), Hebblewhite, M., Merrill, E.H., 2007. Multiscale wolf predation risk for elk: does mi- 1267–1281. gration reduce risk? Oecologia 152 (2), 377–387. Middleton, A.D., Kauffman, M.J., McWhirter, D.E., Jimenez, M.D., Cook, R.C., Cook, J.G., Hebblewhite, M., White, C.A., Nietvelt, C.G., McKenzie, J.A., Hurd, T.E., Fryxell, J.M., Albeke, S.E., Sawyer, H., White, P.J., 2013. Linking anti-predator behaviour to prey
9 R.A. Montgomery, et al. Biological Conservation 233 (2019) 1–11
demography reveals limited risk effects of an actively hunting large carnivore. Ecol. Syst. 28 (1), 289–316. Lett. 16 (8), 1023–1030. Preisser, E.L., Bolnick, D.I., Benard, M.F., 2005. Scared to death? The effects of in- Miller, J.R., Ament, J.M., Schmitz, O.J., 2014. Fear on the move: predator hunting mode timidation and consumption in predator–prey interactions. Ecology 86 (2), 501–509. predicts variation in prey mortality and plasticity in prey spatial response. J. Anim. Preisser, E.L., Orrock, J.L., Schmitz, O.J., 2007. Predator hunting mode and habitat do- Ecol. 83 (1), 214–222. main alter nonconsumptive effects in predator–prey interactions. Ecology 88 (11), Mitchell, W.A., Lima, S.L., 2002. Predator-prey shell games: large-scale movement and its 2744–2751. implications for decision-making by prey. Oikos 99 (2), 249–259. Prugh, L.R., Stoner, C.J., Epps, C.W., Bean, W.T., Ripple, W.J., Laliberte, A.S., Brashares, Mlot, C., 2018. Classic wolf-moose study to be restarted on Isle Royale. Science 61 (6409), J.S., 2009. The rise of the mesopredator. Bioscience 59 (9), 779–791. 1298–1299. Prugh, L.R., Sivy, K.J., Mahoney, P., Ganz, T., Ditmer, M., van de Kerk, M., Gilbert, S., Moll, R.J., Steel, D., Montgomery, R.A., 2016. AIC and the challenge of complexity: a case Montgomery, R.A., 2019. Designing predation risk studies for improved inference in study from ecology. Stud. Hist. Phil. Sci. 60, 35–43. carnivore-ungulate systems. Biol. Conserv. Moll, R.J., Redilla, K.M., Mudumba, T., Muneza, A.B., Gray, S.M., Abade, L., Hayward, Räikkönen, J., Vucetich, J.A., Peterson, R.O., Nelson, M.P., 2009. Congenital bone de- M.W., Millspaugh, J.J., Montgomery, R.A., 2017. The many faces of fear: a synthesis formities and the inbred wolves (Canis lupus) of Isle Royale. Biol. Conserv. 142 (5), of the methodological variation in characterizing predation risk. J. Anim. Ecol. 86 1025–1031. (4), 749–765. Ricklefs, R.E., 2008. The Economy of Nature. Macmillan. Montgomery, R.A., Vucetich, J.A., Peterson, R.O., Roloff, G.J., Millenbah, K.F., 2013. The Riginos, C., 2015. Climate and the landscape of fear in an African savanna. J. Anim. Ecol. influence of winter severity, predation and senescence on moose habitat use. J. Anim. 84 (1), 124–133. Ecol. 82 (2), 301–309. Ripple, W.J., Wirsing, A.J., Wilmers, C.C., Letnic, M., 2013. Widespread mesopredator Montgomery, R.A., Vucetich, J.A., Roloff, G.J., Bump, J.K., Peterson, R.O., 2014. Where effects after wolf extirpation. Biol. Conserv. 160, 70–79. wolves kill moose: the influence of prey life history dynamics on the landscape Ripple, W.J., Estes, J.A., Beschta, R.L., Wilmers, C.C., Ritchie, E.G., Hebblewhite, M., ecology of predation. PLoS One 9 (3), e91414. Berger, J., Elmhagen, B., Letnic, M., Nelson, M.P., Schmitz, O.J., 2014. Status and Nelson, E.H., Matthews, C.E., Rosenheim, J.A., 2004. Predators reduce prey population ecological effects of the world's largest carnivores. Science 343 (6167), 1241484. growth by inducing changes in prey behavior. Ecology 85 (7), 1853–1858. Ripple, W.J., Abernethy, K., Betts, M.G., Chapron, G., Dirzo, R., Galetti, M., Levi, T., Nicholson, A.J., Bailey, V.A., 1935. The balance of animal populations.—Part I. Proc. Lindsey, P.A., Macdonald, D.W., Machovina, B., Newsome, T.M., 2016. Bushmeat Zool. Soc. London 105 (3), 551–598 (Oxford, UK: Blackwell Publishing Ltd, hunting and extinction risk to the world's mammals. R. Soc. Open Sci. 3 (10), 160498. September). Ritchie, E.G., Johnson, C.N., 2009. Predator interactions, mesopredator release and bio- Nilsen, E.B., Linnell, J.D., Odden, J., Andersen, R., 2009. Climate, season, and social diversity conservation. Ecol. Lett. 12 (9), 982–998. status modulate the functional response of an efficient stalking predator: the Eurasian Roemer, G.W., Gompper, M.E., Van Valkenburgh, B., 2009. The ecological role of the lynx. J. Anim. Ecol. 78 (4), 741–751. mammalian mesocarnivore. Bioscience 59 (2), 165–173. Northfield, T.D., Barton, B.T., Schmitz, O.J., 2017. A spatial theory for emergent multiple Rosindell, J., Hubbell, S.P., He, F., Harmon, L.J., Etienne, R.S., 2012. The case for eco- predator–prey interactions in food webs. Ecol. Evol. 7 (17), 6935–6948. logical neutral theory. Trends Ecol. Evol. 27 (4), 203–208. Ogada, M.O., Woodroffe, R., Oguge, N.O., Frank, L.G., 2003. Limiting depredation by Say-Sallaz, E., Chamaillé-Jammes, S., Fritz, H., Valeix, M., 2019. Non-consumptive effects African carnivores: the role of livestock husbandry. Conserv. Biol. 17 (6), 1521–1530. of predation in large terrestrial mammalian systems: mapping our knowledge and Pace, M.L., Cole, J.J., Carpenter, S.R., Kitchell, J.F., 1999. Trophic cascades revealed in revealing the tip of the iceberg. Biol. Conserv (This issue, Under review). diverse ecosystems. Trends Ecol. Evol. 14 (12), 483–488. Schaller, G.B., Vasconcelos, J.M.C., 1978. Jaguar predation on capybara. Z. Säugetierkd. Paine, R.T., 1966. Food web complexity and species diversity. Am. Nat. 100 (910), 65–75. 43, 296–301. Paine, R.T., 1980. Food webs: linkage, interaction strength and community infrastructure. Scheel, D., 1993. Profitability, encounter rates, and prey choice of African lions. Behav. J. Anim. Ecol. 49 (3), 667–685. Ecol. 4 (1), 90–97. Pangle, K.L., Peacor, S.D., Johannsson, O.E., 2007. Large nonlethal effects of an invasive Schmidt, K., Kuijper, D.P., 2015. A “death trap” in the landscape of fear. Mamm. Res. 60 invertebrate predator on zooplankton population growth rate. Ecology 88 (2), (4), 275–284. 402–412. Schmitt, M.H., Stears, K., Shrader, A.M., 2016. Zebra reduce predation risk in mixed- Paquet, P.C., 1992. Prey use strategies of sympatric wolves and coyotes in Riding species herds by eavesdropping on cues from giraffe. Behav. Ecol. 27 (4), 1073–1077. Mountain National Park, Manitoba. J. Mammal. 73 (2), 337–343. Schmitz, O.J., 1998. Direct and indirect effects of predation and predation risk in old-field Parrish, J.K., 1993. Comparison of the hunting behavior of four piscine predators at- interaction webs. Am. Nat. 151 (4), 327–342. tacking schooling prey. Ethology 95 (3), 233–246. Schmitz, O.J., 2003. Top predator control of plant biodiversity and productivity in an old- Patterson, B.R., Benjamin, L.K., Messier, F., 1998. Prey switching and feeding habits of field ecosystem. Ecol. Lett. 6 (2), 156–163. eastern coyotes in relation to snowshoe hare and white-tailed deer densities. Can. J. Schmitz, O.J., 2005. Behavior of predators and prey and links with population-level Zool. 76 (10), 1885–1897. processes. In: Barbosa, P., Castellanos, I. (Eds.), Ecology of Predator–prey Pays, O., Benhamou, S., Helder, R., Gerard, J.F., 2007. The dynamics of group formation Interactions. Oxford University Press, Oxford, UK, pp. 256–278. in large mammalian herbivores: an analysis in the European roe deer. Anim. Behav. Schmitz, O.J., 2008. Effects of predator hunting mode on grassland ecosystem function. 74 (5), 1429–1441. Science 319 (5865), 952–954. Peacor, S.D., 2003. Phenotypic modifications to conspecific density arising from preda- Schmitz, O.J., Beckerman, A.P., O'Brien, K.M., 1997. Behaviorally mediated trophic tion risk assessment. Oikos 100 (2), 409–415. cascades: effects of predation risk on food web interactions. Ecology 78 (5), Peacor, S.D., Werner, E.E., 1997. Trait-mediated indirect interactions in a simple aquatic 1388–1399. food web. Ecology 78 (4), 1146–1156. Schmitz, O.J., Hambäck, P.A., Beckerman, A.P., 2000. Trophic cascades in terrestrial Peacor, S.D., Werner, E.E., 2001. The contribution of trait-mediated indirect effects to the systems: a review of the effects of carnivore removals on plants. Am. Nat. 155 (2), net effects of a predator. Proc. Natl. Acad. Sci. 98 (7), 3904–3908. 141–153. Peacor, S.D., Peckarsky, B.L., Trussell, G.C., Vonesh, J.R., 2013. Costs of predator-induced Schmitz, O.J., Krivan, V., Ovadia, O., 2004. Trophic cascades: the primacy of trait- phenotypic plasticity: a graphical model for predicting the contribution of non- mediated indirect interactions. Ecol. Lett. 7 (2), 153–163. consumptive and consumptive effects of predators on prey. Oecologia 171 (1), 1–10. Schmitz, O.J., Miller, J.R., Trainor, A.M., Abrahms, B., 2017. Toward a community Pearson, R.G., Dawson, T.P., 2003. Predicting the impacts of climate change on the dis- ecology of landscapes: predicting multiple predator–prey interactions across geo- tribution of species: are bioclimate envelope models useful? Glob. Ecol. Biogeogr. 12 graphic space. Ecology 98 (9), 2281–2292. (5), 361–371. Schoener, T.W., 1974. Resource partitioning in ecological communities. Science 185 Peckarsky, B.L., McIntosh, A.R., 1998. Fitness and community consequences of avoiding (4145), 27–39. multiple predators. Oecologia 113 (4), 565–576. Sih, A., Crowley, P., McPeek, M., Petranka, J., Strohmeier, K., 1985. Predation, compe- Peckarsky, B.L., Cowan, C.A., Penton, M.A., Anderson, C., 1993. Sublethal consequences tition, and prey communities: a review of field experiments. Annu. Rev. Ecol. Syst. 16 of stream-dwelling predatory stoneflies on mayfly growth and fecundity. Ecology 74 (1), 269–311. (6), 1836–1846. Sih, A., Englund, G., Wooster, D., 1998. Emergent impacts of multiple predators on prey. Peckarsky, B.L., Abrams, P.A., Bolnick, D.I., Dill, L.M., Grabowski, J.H., Luttbeg, B., Trends Ecol. Evol. 13 (9), 350–355. Orrock, J.L., Peacor, S.D., Preisser, E.L., Schmitz, O.J., Trussell, G.C., 2008. Revisiting Sober, E., 2015. Ockham's Razors: A User's Manual. Cambridge University Press, the classics: considering nonconsumptive effects in textbook examples of pre- Cambridge, England. dator–prey interactions. Ecology 89 (9), 2416–2425. Steenweg, R., Hebblewhite, M., Gummer, D., Low, B., Hunt, B., 2016. Assessing potential Peers, M.J.L., Majchrzak, Y.N., Neilson, E., Lamb, C.T., Hämäläinen, A., Haines, J.A., habitat and carrying capacity for reintroduction of plains bison (Bison bison bison) in Garland, L., Doran-Myers, D., Broadley, K., Boonstra, R., Boutin, S., 2018. Banff National Park. PLoS One 11 (2), e0150065. Quantifying fear effects on prey demography in nature. Ecology 99, 1716–1723. Taylor, R., 1984. Predation Chapman & Hall: New York. New York, USA. Peterson, R.O., Page, R.E., 1988. The rise and fall of Isle Royale wolves, 1975–1986. J. Thaker, M., Vanak, A.T., Owen, C.R., Ogden, M.B., Niemann, S.M., Slotow, R., 2011. Mammal. 69 (1), 89–99. Minimizing predation risk in a landscape of multiple predators: effects on the spatial Petrunenko, Y.K., Montgomery, R.A., Seryodkin, I.V., Zaumyslova, O.Y., Miquelle, D.G., distribution of African ungulates. Ecology 92 (2), 398–407. Macdonald, D.W., 2016. Spatial variation in the density and vulnerability of pre- Tilman, D., 1987. The importance of the mechanisms of interspecific competition. Am. ferred prey in the landscape shape patterns of Amur tiger habitat use. Oikos 125 (1), Nat. 129 (5), 769–774. 66–75. Tollrian, R., Harvell, C.D., 1999. The Ecology and Evolution of Inducible Defenses. Pimm, S.L., 1982. Food webs. In: Food Webs. Springer, Dordrecht, pp. 1–11. Princeton University Press. Polis, G.A., Strong, D.R., 1996. Food web complexity and community dynamics. Am. Nat. Tukey, J.W., 1961. Discussion, emphasizing the connection between analysis of variance 147 (5), 813–846. and spectrum analysis. Technometrics 3 (2), 191–219. Polis, G.A., Anderson, W.B., Holt, R.D., 1997. Toward an integration of landscape and Turner, A.M., 1996. Freshwater snails alter habitat use in response to predation. Anim. food web ecology: the dynamics of spatially subsidized food webs. Annu. Rev. Ecol. Behav. 51 (4), 747–756.
10 R.A. Montgomery, et al. Biological Conservation 233 (2019) 1–11
Turner, A.M., Mittelbach, G.G., 1990. Predator avoidance and community structure: in- carnivores. Ecology 94 (11), 2619–2631. teractions among piscivores, planktivores, and plankton. Ecology 71 (6), 2241–2254. Volterra, V., 1926. Variazioni e Fluttuazioni del Numero d” Individui in specie Animali Turner, A.M., Montgomery, S.L., 2003. Spatial and temporal scales of predator avoidance: Conviventi. Memoire della R. Accademia Nazionale dei Lincei, anno CCCCXXIII, II. experiments with fish and snails. Ecology 84 (3), 616–622. 1926. (Fluctuations in the Abundance of a Species Considered Mathematically). Valeix, M., Loveridge, A.J., Chamaillé-Jammes, S., Davidson, Z., Murindagomo, F., Fritz, Nature 118, 558–560. H., Macdonald, D.W., 2009a. Behavioral adjustments of African herbivores to pre- Werner, E.E., Mittelbach, G.G., 1981. Optimal foraging: field tests of diet choice and dation risk by lions: spatiotemporal variations influence habitat use. Ecology 90 (1), habitat switching. Am. Zool. 21 (4), 813–829. 23–30. Werner, E.E., Peacor, S.D., 2006. Lethal and nonlethal predator effects on an herbivore Valeix, M., Fritz, H., Loveridge, A.J., Davidson, Z., Hunt, J.E., Murindagomo, F., guild mediated by system productivity. Ecology 87 (2), 347–361. Macdonald, D.W., 2009b. Does the risk of encountering lions influence African her- Werner, E.E., Gilliam, J.F., Hall, D.J., Mittelbach, G.G., 1983. An experimental test of the bivore behaviour at waterholes? Behav. Ecol. Sociobiol. 63 (10), 1483–1494. effects of predation risk on habitat use in fish. Ecology 64 (6), 1540–1548. Valeix, M., Loveridge, A.J., Davidson, Z., Madzikanda, H., Fritz, H., Macdonald, D.W., Winemiller, K.O., Polis, G.A., 1996. Food webs: what can they tell us about the world? In: 2010. How key habitat features influence large terrestrial carnivore movements: Food Webs. Springer, Boston, MA, pp. 1–22. waterholes and African lions in a semi-arid savanna of north-western Zimbabwe. Wirsing, A.J., Heithaus, M.R., Dill, L.M., 2007. Fear factor: do dugongs (Dugong dugon) Landsc. Ecol. 25 (3), 337–351. trade food for safety from tiger sharks (Galeocerdo cuvier)? Oecologia 153 (4), Valeix, M., Chamaillé-Jammes, S., Loveridge, A.J., Davidson, Z., Hunt, J.E., Madzikanda, 1031–1040. H., Macdonald, D.W., 2011. Understanding patch departure rules for large carni- Wolf, C., Ripple, W.J., 2016. Prey depletion as a threat to the world's large carnivores. R. vores: lion movements support a patch-disturbance hypothesis. Am. Nat. 178 (2), Soc. Open Sci. 3 (8), 160252. 269–275. Woodroffe, R., Thirgood, S., Rabinowitz, A., 2005. People and Wildlife, Conflict or Co- Vanak, A.T., Fortin, D., Thaker, M., Ogden, M., Owen, C., Greatwood, S., Slotow, R., 2013. existence? (No. 9). Cambridge University Press. Moving to stay in place: behavioral mechanisms for coexistence of African large
11